DESCRIPTION (Applicant's Abstract): Effective social interaction and personal adaptation requires that people often try to influence their own minds. People engage in such mental control when they try to stop thinking about something, for example, or when they attempt to avoid unwanted or inappropriate emotions or desires. Attempts to control mental states often fail, however, and past research suggests that this happens because mental control produces ironic processes unconscious failure-monitoring processes that promote the appearance of the mental states people most hope to avoid. This is why unwanted thoughts and emotions often arise at precisely the worst times. The specific aim of this project is to see whether such ironic processes are overcome when mental control is practiced repeatedly and becomes automatized Like any practiced skill, mental control should become more effective with rehearsal, and this research examines its possibility by focusing both on rehearsal of mental control in the laboratory and on the testing of special populations who are likely to have engaged in chronic mental control in the past. But also like any skill that has become automatic, mental control can be disrupted whenever the person renews conscious control and monitoring of the process. This fragile automaticity hypothesis is tested in these studies, to see whether remembering that one is trying to control one's mind can reintroduce ironic thoughts and emotions even after mental control has become automatic. This research should have several key implications for mental health. The automatization of mental control processes, and its accompanying fragile automaticity, may be implicated in the etiology of depression, obsessive-compulsive disorder, and eating disorders, and may also underlie social interfactional problems that arise from error in the control of socially inappropriate behavior.